Professional Documents
Culture Documents
10 1 1 948 6664ddddddinversmatrice
10 1 1 948 6664ddddddinversmatrice
Abstract. In the paper certain examples of applications of the matrix inverses for generat-
ing and calculating the integrals are presented.
Introduction
The first part of our discussion concerns the linear mappings defined on the
finite-dimensional space of solutions of the following system of differential equa-
tions
= , + ⋯ + ,
= , + ⋯ + ,
(1)
⋯ + , .
⋮
= , +
Matrix is nonsingular, so there exists its inverse . In particular, the follow-
ing equality occurs:
= ⋮ .
⋮
(3)
64 D. Matlak, J. Matlak, D. Słota, R. Wituła
Therefore, if = ⋮ , then functions are the primitive functions
⋮
of ( = 1,2, … , ).
The current article is inspired by Swartz’s paper [1] where the author gives
integrals of functions ℎ = e
sin , ℎ =
cos , for which he obtained the
some simple examples of using this procedure, among others, for generating the
formula (the integration constants are omitted and this rule will oblige hencefor-
ward):
ℎ # − $%#
= ! = మ
మ " &.
ℎ ℎ # + $%#
− ℎ ೌೣ
(4)
Let us start from the generalization of the example mentioned above. Let us take
the functions
Note that the differentiation operator for these functions is of the form
' ( = * , = - . ' (.
0 0
−
− 0 0
(6)
) +
0 0
0 0
= - ..
0 − 0
0 0
(7)
మ
మ 0 0
0 − 0
−
+
/
=
+
.
Differentiation and integration by using matrix inversion 65
sin
' (=* ,=' (=
sin
(9)
sin
−1
=- .' (.
0 ⋯ 0 0
3 ∙ 2 −3 ⋯ 0 0
⋯ 00 − 1
−0 sin
⋮ ⋮ ⋱ ⋮ ⋮ ⋮
0 0
= , ೖశభ×ೖశభ , (10)
మ మ
where
< 4,
= 4,
0,
, =
‼ ‼
− మ ,
> 4.
(11)
− ‼ ‼ ,
We can deduce that for odd there occurs (with respect to the linear element):
for = 2,4,6, … . Acting on the vector ' (, where 0 is even, with the second
⋮
derivative operator, like it was done in equation (9), we get the following transfor-
mation matrix:
6 ' ( = ' (=
⋮ ⋮
(16)
2 ∙ 3⁄4 −4 …
=- .' (.
−2 0 … 0 0
0 0
… 0 − 2
0 − 1
⁄0 −0
⋮ ⋮ ⋱ ⋮ ⋮ ⋮
0 0
The above matrix is invertible and its inversion is of the form
6 = , ೖ×ೖ, (17)
< 4,
మ మ
= 4,
0,
, =
− మ ,
> 4.
where (18)
− మ ‼ ‼ ,
‼ ‼
Therefore for even n we get the formula (exact to the linear element):
" sin − 1 +
‼మ ‼
‼ ‼
= −
− మ sin = − మ sin ,
‼
= ‼ మ
Differentiation and integration by using matrix inversion 67
4. Integral of ࢞ ܖ܉ܜ
tan
tan
tan
' ( = < ' (.
1
tan
tan
(23)
⋮ ⋮
If < is represented by infinite matrix , then from (23) matrix has the form
=- ..
1 0 1 0 0 ⋮
0 2 0 2 0 ⋮
(24)
0 0 3 0 3 ⋮
⋯ ⋯ ⋯ ⋯ ⋯ ⋯
It is easy to show that
?1 0 − 0 ⋮B
> ⋮ A.
0
= >0 A
>0 0 0 − ⋮ A
0 − 0 (25)
=⋯ ⋯ ⋯ ⋯ ⋯ ⋯@
However, matrix does not represent the inverse operator < , since the fol-
lowing relations hold
68 D. Matlak, J. Matlak, D. Słota, R. Wituła
tan
tan
tan
' (=- . ' (, (26)
1 0 0 0 −1 0 0 0 ⋮ 1
tan
tan
0 1 0 0 0 −1 0 0 ⋮
0 0 1 0 0 0 −1 0 ⋮
⋮ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋯ ⋮
resulting from the formula
Moreover, we get from this, after summing over powers in the interval − , and
by uniform convergence (see [4]), that
tan =
∑
" tan − tan &,
(28)
or equivalently
tan CC = ∑
" tan tan &
−
= ∑ ,
ೖ
(29)
tan
From formula (28) we also get the matrix form D<
of operator < , i.e. the
inverse operator of operator < (under assumption of its existence):
? B
1 0 0 0 0 ⋯
> 0 ⋯A
>− 0 0 ⋯A
0
0 0
D<
= > A.
0
>0 − 0 ⋯A
(30)
> A
0
> A
= ⋮ ⋮ ⋱@
0 − 0
⋯
⋮ ⋮ ⋮
Formula (29) is our main analytic result in this section. Why do we think so?
Because, as we show now, this formula is a generalization of the classical
MacLaurin’s formulae for ln( + 1), i.e.
ln + 1
= −
మ
య
ర
+ − + ⋯, (31)
for −1 < ≤ 1, found independently by Nicolaus Mercator and Saint-Vincent
(see sections 10-9 and 10-10 in [5] and page 387 in [6]), and for arctan , i.e.
arctan = −
య
ఱ
+ − ⋯, (32)
for −1 < ≤ 1, which is known as the Gregory series.
Differentiation and integration by using matrix inversion 69
This connection should not be surprising because of the known complex rela-
tion (see [7]):
where E⁄ ∈
(−∞, −1] ∪ [1, ∞) and where the principal branch of the loga-
rithm is under consideration. On the cuts we have
for C ∈ (−∞, −1) ∪ (1, ∞) and where the upper/lower sign corresponds to the
right/left side of the set determining C. More precisely, the connection between the
where E = + C, |E| < 1. First, from equation (29) for = 1 we get
− ln cos = ∑
ೖ
()
tan (36)
or
ln cos1arctan √2 = ∑
ೖ
, (37)
ln cos1arctan √2 = − ln + 1
,
(38)
i.e.
(29) we get
= ∑
ೖ
!$% !"
&
మ '
, (40)
70 D. Matlak, J. Matlak, D. Słota, R. Wituła
As seen from equation (29), the values of integrals ర tan for ≥ 2 are
tions of the Gregory power series (32) are discussed in papers [9] and [10].
ഏ
the translations of numbers ర tan or ర , depending on parity of . For
ഏ ഏ
even we have
5. Final remark
sec , similarly as in the present study. The obtained formulae were used
paper [12] the technique of the inverse matrix was used for calculating the integral
References
[1] Swartz W., Integration by matrix inversion, Amer. Math. Monthly 1958, 65, 282-283.
[2] Prudnikov A.P., Bryczkov J.A., Mariczev O.J., Integrals and Series, Elementary Functions,
Vol. 1, Nauka Press, Moscov 1986 (in Russian).
[3] Prudnikov A.P., Bryczkov J.A., Mariczev O.J., Integrals and Series, Complements Sections,
Vol. 2, Nauka Press, Moscow 1986 (in Russian).
[4] Kołodziej W., Mathematical Analysis, PWN, Warsaw 1979 (in Polish).
[5] Eves H., An Introduction to the History of Mathematics, Holt, Rinehart and Winston, New York
1969.
[6] Boyer C. (revised by Merzbach U.), A History of Mathematics, Wiley, New York 1991.
[7] Olver F.W.J., Lozier D.W., Boisvert R.F., Clark C.W., NIST Handbook of Mathematical Func-
tions, Cambridge Univ. Press, Cambridge 2010.
[8] Wituła R., Number π, its History and Influence on Mathematical Creativeness, Wyd. Pracowni
Komputerowej Jacka Skalmierskiego, Gliwice 2011 (in Polish).
Differentiation and integration by using matrix inversion 71
[9] Gawrońska N., Słota D., Wituła R., Zielonka A., Some generalizations of Gregory’s power
series and their applications, J. Appl. Math. Comput. Mech. 2013, 12(3), 79-91.
[10] Wituła R., Hetmaniok E., Pleszczyński M., Słota D., Generalized Gregory’s series (in review).
[11] Redheffer R., Induced transformations of the derivative - vector, Amer. Math. Monthly 1976,
83, 255-259.
[12] Wituła R., Matlak D., Matlak J., Słota D., Use of matrices in evaluation of sec ଶାଵ ݔ݀ ݔ
in review.